Adrenocortical Steroid Metabolism and Adrenal Cortical Function in Liver Disease

Home , Dihydrocortisone, Tetrahydrocorticosterone

ADRENOCORTICAL STEROID METABOLISM AND ADRENAL CORTICAL FUNCTION IN LIVER DISEASE

Ralph E. Peterson

J Clin Invest. 1960;39(2):320-331. https://doi.org/10.1172/JCI104043.

Research Article

Find the latest version: https://jci.me/104043/pdf ADRENOCORTICAL STEROID METABOLISM AND ADRENAL CORTICAL FUNCTION IN LIVER DISEASE BY RALPH E. PETERSON * (From The National Institute of Arthritis and MVetabolic Diseases, Bethesda, Md.) (Submitted for publication July 27, 1959; accepted October 1, 1959) Zondek (1) as early as 1934 demonstrated that to man disappear rapidly from the blood (15, 16). enzymes of the liver destroyed the biological ac- Only minimal quantities are lost via the expired tivity of the estrogens. Since then both in tivo CO2 (17) or the biliary or gastrointestinal tract and in vitro studies have provided much evidence (15, 16). Also, practically all of the administered to show that the liver is the organ primarily re- steroid is metabolized prior to its excretion in the sponsible for the catabolism of the steroid hor- urine (15, 16), and thus it is apparent that urinary mones: estrogens, androgens, progesterone, and excretion plays a relatively minor part in elimi- the corticosteroids (2). However, not until the nating the biologically active steroid. Thus, meta- development of improved methods for measure- bolic transformation by the liver must play the pre- ment of certain of the adrenocortical steroids and dominant role in terminating the action of the their metabolites, and the availability of labeled steroids. radioactive cortisol, corticosterone, and aldosterone Because of the major role of the liver in the has it been possible accurately to evaluate the in- catabolism of the corticosteroids, it might be ex- fluence of liver disease on the rate of degradation pected that this organ could indirectly influence and synthesis of the steroids in man. The studies the synthesis of corticosteroids by the adrenals. here reported are based on the use of certain of Many investigators have attempted to determine these newer techniques. whether alterations in the functional capacity of the On incubation with rat liver tissue, cortisol and liver result in changes in adrenal cortical function. cortisone are rapidly metabolized, but only very Until recently studies of adrenal cortical function slowly with other tissues (3-5). Perfusion stud- in both acute and chronic liver disease were based ies have demonstrated a very rapid metabolism of on measurements of urinary steroids. There has the steroids by the liver but not by other organs been general agreement that the urinary 17-ke- (6-8). Hechter, Frank, Caspi and Frank (9) tosteroids are low in various forms of liver disease found that the major portion of the cortisone and -acute viral hepatitis (18-21), toxic hepatitis cortisol administered into the portal vein in dogs (20), portal cirrhosis (18-22), biliary cirrhosis was not recovered as unaltered steroid from the (23, 24), and also in obstructive jaundice (18-21, hepatic venous blood. Bradlow, Dobriner and 25). Urinary androgen excretion, as measured by Gallagher (10) found that 70 per cent of a dose of bioassay, has also been found to be low in cirrhosis tritium-labeled cortisone administered to mice of the liver (26, 27). However, since only a small was found in the liver within five minutes after fraction (5 to 10 per cent) of the cortisol is me- intravenous administration. Administered cortisol tabolized to 17-ketosteroids (28), the urinary also disappears rapidly from the circulation in rats level of this material does not represent an ade- (11), and this rapid metabolism can be prevented quate index of the functional capacity of the by hepatectomy but not by nephrectomy (12). adrenal cortex to secrete corticosteroids such as The liver in man has a high capacity for metabo- cortisol. Furthermore, the rate of adrenal cortical lizing the circulating blood cortisol, as demon- secretion of 17-ketosteroids may not always paral- strated by the fact that the level of 17-hydroxy- lel the rate of secretion of adrenal corticosteroids. corticosteroids in the hepatic vein blood is lower Although urine levels of "corticoids" have been than the level in the arterial blood (13, 14). reported to be normal or elevated in acute hepatitis Adrenocortical steroids administered intravenously (21) and cirrhosis (21, 23, 25, 29-31), the data are difficult to interpret because * Now at Cornell University Medical College, New in these studies York, N. Y. of the nonspecific methods of assay used. Brown, 320 ADRENAL FUNCTION IN LIVER DISEASE 321

Willardson, Samuels and Tyler (32), using a more specificity of this assay method for each steroid was eval- these ster- reliable assay method, found the urinary corticoids uated by isotope dilution (37).2 For each of oids, the isotope dilution assay indicated that at least 80 to be low in cirrhosis. Plasma 17-hydroxycorti- per cent of the steroid as measured by the phenylhydra- costeroid levels have been reported to be normal zine assay represented the administered steroid. in cirrhosis of the liver (33, 34). In other stud- The rate of disappearance of cortisol4-C'4 from the ies of cortisol metabolism in liver disease, infused plasma and its rate of appearance in the ascitic fluid cortisol was reported to disappear from the plasma was determined by the double labeling technique using cortisol-H' (42). The rate of disappearance of aldoster- at a decreased rate in patients with hepatitis or one from the plasma after infusion of aldosterone-H' was cirrhosis of the liver (15, 32, 33). determined by the double labeling technique using al- Studies of the adrenals in patients with cirrhosis dosterone and acetic-i-C14 anhydride (42). of the liver coming to autopsy have demonstrated Cortisone and corticosterone were determined by the a decrease in lipoid material (35) and narrow and isotope dilution method, since the phenylhydrazine assay for cortisone (43) and, to a lesser extent, the fluoro- frequently nodular adrenal cortices (36). metric assay for corticosterone (40) were not found to be specific for the determination of these steroids in MATERIALS AND METHODS plasma following their intravenous injection. Fourteen patients with moderately severe cirrhosis of Urine cortisol concentration following the infusion of the liver, as judged by clinical signs and symptoms and cortisol was determined by isotope dilution. Following liver function tests, and 3 patients with acute viral hepa- the injection of cortisol4-C'4, the total radioactivity in titis, served as the subjects of this study. Those with the urine was determined by counting a small aliquot of cirrhosis of the liver had the classical symptoms, signs, the urine (0.1 to 0.5 ml) in an alcohol-toluene, DPO and laboratory findings, viz. prolonged sulfobromoph- (diphenyloxyzol) and POPOP [1,4-bis-2- (5-phenyloxa- thalein retention, reversed albumin/globulin ratio with zolyl)-benzene] phosphor in the liquid scintillation spec- low serum albumin, elevated serum bilirubin, and ab- trometer. The fractions of radioactive metabolites ap- normal thymol turbidity or cephalin flocculation tests. pearing as free and glucuronide conjugates were de- On clinical grounds, most of these patients were presumed termined by previously described methods (15). The to have alcoholic cirrhosis, and in the majority of the urine concentrations of tetrahydrocortisol following in- subjects the diagnosis of cirrhosis was confirmed by nene-3, 20-dione). Aldosterone (11j, 21-dihydroxy4- liver biopsy. One of the 3 patients with acute hepatitis pregnene-3, 20-dione-18, al). Dihydrocortisone (17a, 21- had homologous serum (post-transfusion) hepatitis. dihydroxy-pregnane-3, 11, 20-trione). Dihydrocortisol Plasma cortisol and urinary corticosteroids were de- (ll,B, 17a, 21-trihydroxy-pregnane-3, 20-dione). Tetra- termined by the modified (37) procedure of Silber and hydrocortisone (3a, 17a, 21-trihydroxy-pregnane-11, 20- Porter (38) using phenylhydrazine and sulfuric acid. dione). Tetrahydrocortisol (3a, llfi, 17a, 21-tetrahy- Urinary 17-ketosteroids were determined by the Zimmer- droxy-pregnane-20-one). Tetrahydrocorticosterone (3a, mann procedure, modified from the Holtorff and Koch 11,, 21-trihydroxy-pregnane-20-one). 9a-Fluorocortisol method (39). Plasma corticosterone was determined by (9a-fluoro-ll,8,-7a, 21-trihydroxy4-pregnene-3, 20-di- an isotope dilution method (40). Aldosterone excretion one). A' 9a-Fluorocortisol (9a-fluoro-iip, 17a, 21-tri- in the urine was determined by the double isotope deriva- hydroxy-1,4-pregnene-3, 20-dione). tive method (41, 42). 2 mc for intravenous adminis- Cortisone-4-C'4 (0.49 per mmole), cortisol-4-C' The steroids (100 to 200 mg) (1.47 mc per mmole), corticosterone4-C'4 (1.47 mc per tration were dissolved in ethanol and diluted to 300 ml mmole), tetrahydrocortisone-4-C'4 monoacetate (0.4 mc per dextrose in water to a final alcohol con- with 5 cent per mmole), and dihydrocortisone-4-C' acetate (0.4 mc centration of approximately 5 per cent. Following of the steroids between per mmole) were obtained from the Endocrine Study rapid (10 to 15 minutes) infusion Section of the National Institutes of Health. The acetates 8 and 9 a.m., plasma samples were collected serially at 20 were converted to the free steroids by hydrolysis with for 2 to 3 hours and assayed. to 30 minute intervals acetyl cholinesterase as described previously (43). C'4- Plasma steroid concentration was plotted semilogarithmi- of the labeled tetrahydrocortisol and dihydrocortisol were pre- cally against time, and the rate of disappearance pared enzymatically (44) from cortisol-4-C'4 (1.47 mc per as a half- steroids from the plasma expressed biologic mmole). Cortisol (0.11 ,uc per ,ug), 9a-fluorocortisol time. The following plasma steroids were assayed by the (0.20 ,uc per ,ug), A' 9a-fluorocortisol (0.05 /c per ,ug), phenylhydrazine method: cortisol, dihydrocortisone, di- A' tetrahydrocortisone, tetrahydrocortisol, 9a- cortisol (0.70 /Ac per ,ug), corticosterone (1.5 juc per ,sg), hydrocortisol, and aldosterone per were labeled with tritium A' 9a-fluorocortisol, and A'-cortisol.1 The (1.5 /Ac dg), fluorocortisol, by the Wilzbach (45) technique by a method previously 1 In this paper the trivial names of the steroids have described (42). All of the labeled steroids were purified been used. Cortisol (118, 17a, 21-trihydroxy-4-pregnene- by paper chromatography to a constant specific activity. 3, 20-dione). Cortisone (17a, 21-dihydroxy-4-pregnene-3, All of the radioactivity measurements were made with 11, 20-trione). Corticosterone (118, 21-dihydroxy-4-preg- the Packard liquid scintillation spectrometer. 322 RALPH E. PETERSON

TABLE I Plasma and urine steroid levels

Plasma steroids Urine steroids Cortico- 17-Keto- Age Cortisol sterone Corticoids steroids

Ag/100 ml mg/day E. M. e 38 Cirrhosis 17 J. H. 6' 30 Cirrhosis 11 C. C. 6' 27 Cirrhosis 13 0.9 9.1 D. R. 9 52 Cirrhosis 12 0.4 2.8 0.5 T. T. o' 44 Cirrhosis 16 0.2 1.9 1.0 J. C. 6' 60 Cirrhosis 15 0 2.4 2.0 J. M. ci 51 Cirrhosis 14 4.0 1.9 M. B. 9 35 Cirrhosis 21 0.8 3.0 1.5 E. W. c' 35 Cirrhosis 16 0.4 3.5 8.2 R. N. 9 52 Cirrhosis 9 0 4.5 2.0 A. C. 6' 50 Cirrhosis 6 0.2 1.0 1.8 E. F. 9 60 Cirrhosis 13 1.2 0.8 G. D. 6' 50 Hepatitis 14 3.6 5.5 Control* 15 4.9 7.o C. C. Q 24 Hepatitis 13 1.5 0.5 Control* 18 3.5 10.5 R. W. 6' 27 Hepatitis 10 4.4 11.0 Normals (90)t (20) Range 6-25 0.5-2.0 9 4-10(20) 9 6-15 (18) Mean,SD 13.51:3 1.1:40.4 6 1.2 1041.0

6 4-14 (36) 6' 10-25 (25) 7 =1 1.5 15 4- 1.5 * Four to six months following recovery. t Number of subjects ( ). fusion of 200 mg cortisol were determined by isotope times found to be low or in the low normal range.3 dilution. Glucuronidase hydrolysis of the urine (15) The urinary corticosteroid and 17-ketosteroid was used in measuring the urine concentrations of radio- concentrations were active steroid metabolites conjugated with glucuronic below normal in the great acid after the infusion of tetrahydrocortisone-4-C14. majority of the patients. The size of the rapidly exchangeable miscible pool of Cortisol (100 to 200 mg) was infused into eight cortisol and the rate of turnover of the pool were deter- patients with cirrhosis of the liver and into one mined by the previously described method (46) using patient with hepatitis, and in each subject the in- 0.2 to 0.3 mg cortisol-4-C'4 for intravenous injection. fused steroid disappeared from the plasma at a The pool size and rate of turnover of corticosterone rate slower than normal (Table II). These find- were determined by the procedure used for cortisol, ex- ings agree with the results of a similar study of 12 cept that tritium-labeled corticosterone (7 to 30 Ag) was used. In a few subjects the rates of turnover of corti- patients with cirrhosis reported previously (15). sol, corticosterone, and aldosterone were measured by In both cirrhotic and normal subjects, cortisol determining the dilution of injected labeled steroid by disappeared from the circulation more rapidly one or more of its urinary metabolites (4749). when infused in trace quantities than in pharmacologic (100 to 200 mg) quantities (49, 50). Fig- RESULTS ure 1 shows the results of studies carried out in a Table I lists the plasma cortisol and corticoster- patient with cirrhosis and massive ascites (J.C.) given both a trace and a pharmacologic one levels, and the urine corticosteroid and 17- dose of ketosteroid levels in the patients with liver disease. 3 In three patients with cirrhosis of the liver and as- Plasma cortisol was normal in all cites, maintained on a low salt diet, the plasma aldosterone the subjects; level ranged from 0.08 to 0.15 jug per cent (normal 0.03 however, plasma corticosterone levels were some- to 0.08 ,g per cent). ADRENAL FUNCTION IN LIVER DISEASE 323

TABLE II Biologic half-times of infused steroids (hours)

F* B E DHE DHF THE THF T. T. ci Cirrhosis 5.50 1.58 0.58 0.65 0.75 0.65 0.80 D. R. 9 Cirrhosis 3.08 1.72 0.67 0.50 0.65 0.65 0.75 J. C. c Cirrhosis 4.16 1.33 0.53 0.45 0.45 0.70 0.75 M. B. 9 Cirrhosis 5.00 1.83 0.50 0.60 0.55 J. M. ci Cirrhosis 3.33 R. N. 9 Cirrhosis 3.16 1.55 J. H. ci Cirrhosis 2.67 C. C. c3 Cirrhosis 10.3 1.40 0.50 C. D. oi Hepatitis 3.17 Controlt 2.17 Normals (20)$ (10) (12) (4) (4) (7) (7) Range 1.50-2.00 1.16-1.50 0.4-0.6 0.5-0.8 0.5-0.9 0.5-0.9 0.5-0.8 Mean, SD 1.83 :1: 0.10 1.33 1 0.05 0.45 i 0.03

*B F=- Urnrtican1otiUls eVVre2nMn mo100llmnllfUsUinfiic.- e +v..INmbem£nnfikRofsubeowitgfnRllrJincI r1nVrecovery. B = corticosterone; 100 mg infused. t Number of subjects ( ). E = icortisone;200 mg infused. DHE = dihydrocortisone; 100 mg infused. DHF = dihydrocortisol; 100 mg infused. THE = tetrahydrocortisone; 100 mg infused. THF = tetrahydrocortisol; 100 mg infused. cortisol-4-C04. It is apparent that even though normal subjects metabolized these four compounds the pharmacologic dose of cortisol-4-C14 disap- at much the same rate (Table II). peared from the plasma at a slower rate, there was In one patient with cirrhosis of the liver (T.T.), a more rapid accumulation of a larger quantity of given tritium-labeled aldosterone, the labeled ster- the labeled cortisol in the extravascular fluid (ascitic fluid). With a similar study in Patient T.T. I I I with minimal ascites, infusion of a trace quantity * I mg. Cortisol-4-C14 I.V. (5jc) (0.2 mg) of labeled cortisol gave a biologic half- 50,000 - o200 mg. Cortisol - 4- C14 I.V. ( Stc ) time of 3.0 hours whereas infusion of 200 mg of labeled steroid gave a half-time of 5.5 hours. - - _ tY2sM255Min. Infused cortisone and corticosterone were me- PLASMA tabolized at a rate only slightly slower than normal 0 10,000 _½u240oMin. (Table II). The highly potent synthetic steroids, 0. w _t___ 9a-fluorocortisol and A' 9a-fluorocortisol, were 5,000 metabolized at a normal rate = 1.3 to 1.8 I- (t1/2 z hours) in two patients (T.T. and C.C.) with cir- i rhosis, whereas AW-cortisol was metabolized at a ASCITIC FLUID rate slower (t1/2 = 5.8 hours, T.T.) than normal IL- (t1/2 = 3.0 to 4.0 hours). 2 1I000 0 Four of the major metabolites of cortisol (the di- U hydro and tetrahydro reduction products of cor- 500 tisol and cortisone) were infused in doses of 100 mg. During the period of the study, endogenous plasma cortisol was reduced to zero by the prior I administration of 2.5 mg of A' 9a-fluorocortisol 0 20 40 60 80 100 120 140 orally eight hours before the infusion of the ster- TIME (MINUTES) oids. All of these steroids were metabolized rap- FIG. 1. PLASMA AND ASCITIC FLUID LEVELS OF CORTISOL idly, and both the patients with cirrhosis and the AFTER INTRAVENOUS ADMINISTRATION OF cORTIsoL-4-C14. 324 RALPH E. PETERSON oid disappeared from the plasma at a rate only portant difference in the fraction of the injected slightly slower (tl/2 = 70 minutes) than normal steroid radiometabolites excreted in the urine. A (tl/2= 30 to 60 minutes) whereas in another pa- kinetic study of the appearance of a major glu- tient with severe cirrhosis a half-time of 40 minutes curonide-conjugated metabolite of infused cortisol, was observed. tetrahydrocortisol, however, revealed that the me- Table III lists the concentrations of cortisol in tabolite appeared in the urine at a slower rate in the urine following the infusion of 200 mg corti- the cirrhotic subject (Figure 2), although infused sol to normal subjects and to patients with cir- labeled tetrahydrocortisone appeared in the urine rhosis. A larger fraction of the infused cortisol as glucuronide conjugates at a similar rate in the appeared in the 24 hour urine in the cirrhotic. normal and in the cirrhotic (Figure 3). In Sub- However, both the normal and the cirrhotic groups ject J.D., with familial nonhemolytic icterus, a excreted the same fraction of the infused steroid condition known to be associated with a defect in as cortisol metabolites conjugated as glucuronides the liver enzyme system necessary for the con- and giving the reaction with phenylhydrazine. jugation of many substances with glucuronic acid Following the injection of tracer doses of cortisol- (51), the rate of appearance of the tetrahydrocorti- 4-C14, the normal subjects and the patients with sone glucuronide was delayed (52). cirrhosis excreted essentially the same fraction of Table IV gives the data on the miscible pool the injected steroid as free and as glucuronide- size and the rate of turnover of cortisol in six pa- conjugated metabolites. There was also no im-

I0 500 O L.C. NORMAL LO C * 00 . cE T.T. CIRHHOSIS mgs. 5 TR. Control ZS J. D. NON-HEMOLYTIC ICTERUS 0 o T T Cirrhosis (0.75Lc + 50 mgs). 0. 0 0 100 0o w I w t1 a2.2 hr. i O- w cn- w w 0 4 8 12 t l62OHours CfE ILlI (1) C.) 0 'OF cE t, = 3.1 hr. at\ti z4.2 Hours I

-J -J ~~~~~~~~~~~0 -J 0.1 CE ~~ ~ ~ TM (HOUR2

2 4 6 8 10 12 0.1~~~~~~~~~~~ TIME (Hours) FIG. 3. URINARY EXCRETION OF RADIOACTIVE METABO- 0 4 8 12 1,6 20 24 LITES FOLLOWING INJECTION OF TETRAHYDROCORTISONE- TIME (HOURS) 4-C'4. The radioactive metabolites represent the labeled FIG. 2. URINARY EXCRETION OF TETRAHYDROCORTISOL metabolites of tetrahydrocortisone present in the urine as FOLLOWING INFUSION OF 200 MG CORTISOL. The tetrahy- glucuronide conjugates. The assays were performed on drocortisol measurements represent the tetrahydrocorti- urine fractions that had been previously extracted with di- sol present in the urine as a glucuronide conjugate. The chloromethane to remove the unconjugated steroids. In assays were performed on urine fractions that had been all three subj ects the labeled free steroids represented previously extracted with dichloromethane to remove the less than 3.5 per cent of the administered tetrahydro- unconjugated tetrahydrocortisol. cortisone-4-C'4. ADRENAL FUNCTION IN LIVER DISEASE 325 tients with cirrhosis and in one patient with acute TABLE III hepatitis. In the patient with acute hepatitis the Urinary cortisol and metabolites of cortisol* studies were performed during the acute stage of the disease and again six months following re- 0.2-0.5 mg C14 cortisol (F) iv 200 mg F iv % Total excreted covery. and at a time when all of the liver function % PNHt Excreted Glucu- glucu- tests had returned to normal, although some total Free ronide F ronides hepatomiiegaly persisted. Patients J.M., M.B., and mg G.D. did not have ascites and Patients D.R., C.C., T. T. 80 4.0 50 5.0 40 and T.T. had a moderate ascites. Patient J.C. was D. R. 76 4.5 45 5.6 46 M. G. 82 6.0 60 6.0 48 first studied at a time when he had marked ascites J. C. 70 3.3 50 5.0 38 (20 to 30 L) and one week after removal of 8.5 L of ascitic fluid. The removal of this vol- Normals (8) Normals (5) (8) ume of ascitic fluid did not alter the rate of turn- 74-85 3.8-5.5 50-64 2.5-3.8 40-64 ov-er of cortisol. * These percentages represent the excretion of steroids It may be noted that the fraction of the pool re- during the 24 hour period following their administration. placed per hour was much decreased in the cir- t PNH = phenylhydrazine-reacting steroids. rhotics althouigh the size of the miscible pool of cortisol was usually found to be slightly increased. rate of secretion of cortisol is known to be near When the turnover rate was calculated on the basis maximal and probably greater in magnitude than of the dilution of one of the major metabolites of the average secretion rate (46, 49). the injected cortisol-4-C14, it was also shown that A repeat turnover rate study was carried out in the rate of secretion of cortisol was markedly sup- one patient (T.T.) during the 10 to 14 hour pe- pressed in cirrhosis. When expressed as the rate riod following the continuous intravenous infusion of secretion of cortisol per day (calculated as the of 4 units of corticotropin per hour. The plasma product of the rate of turnover per hour x 24 cortisol level in this patient rose rapidly to a level hours) these two methods of assay were not in of 48 ,tg per cent six hours following the start of agreement. This difference probably results from the infusion and remained at this level during the the fact that the rate studies cover only a three subsequent 18 hours. The pool of cortisol was hotur perio(I of the day (9 a.m. to noon), when the elevated, but the rate of synthesis of cortisol was

TABLE IV Mfiscible pool and rate of turnover of cortisol in liver disease

Pool tt* kt Turnover rate mg hr pools/hr mg/hr$ mg/daA T. T. e Cirrhosis 1.9 6.90 0.10 0.19 1.9 Cirrhosis, ACTH 13.6 5.00 0.14 1.90 55 D. R. 9 Cirrhosis 3.6 8.60 0.08 0.29 4.0 M. B. 9 Cirrhosis 2.5 6.30 0.11 0.27 4.0 R. N. 9 Cirrhosis 2.0 1.78 0.39 0.78 9.0 J. M. c Cirrhosis 1.5 2.57 0.27 0.41 J. C. c' Cirrhosis, ascites 3.5 4.10 0.17 0.59 Cirrhosis, post-ascites 3.3 4.60 0.15 0.49 G. D. ce Hepatitis 1.7 2.77 0.25 0.42 Control 1.5 1.61 0.43 0.65 Normals (14) Range 1.1-2.4 1.0-2.3 0.3-0.7 0.7-1.2 8-20 Mean, SD 1.7 4 0.2 1.4 :=0.2 0.5 + 0.1 0.85 ±i 0.08 14 + 1.5 * Time for specific activity of plasma cortisol to be reduced by 50 per cent. t Fraction of pool replaced per hour. t Calculated from the kinetic method of assay of change in specific activity of plasma cortisol after injection of 200 to 300 Ag of cortisol-4-CI4. § Calculated from specific activity of tetrahydrocortisone in 24 hour urine collected after 200 to 300,g of cortisol- 4-C 4. 326 RALPH E. PETERSON

TABLE V Miscible pool and rate of turnover of corticosterone in liver disease

Pool tj k Turnover rate mg hr pools/hr mg/hr* mg/dayt T. T. Cirrhosis 0.18 1.54 0.45 0.08 0.6 D. R. Cirrhosis 0.30 2.10 0.33 0.10 1.0 M. B. Cirrhosis 0.11 1.10 0.63 0.07 0.3 Normals (11) Range 0.2-0.4 0.8-1.5 0.4-0.9 0.12-0.22 1.5-4.0 Mean, SD 0.3 :41 0.04 1.1 4 0.1 0.6 := 0.1 0.15 + 0.02 2.1 + 0.2 * Calculated from the kinetic method of assay of change in specific activity of plasma corticosterone after injection of 7 to 30,ug of corticosterone-H3. t Calculated from specific activity of tetrahydrocorticosterone in 24 hour urine collected after 7 to 30 ;tg of corticosterone-H3. less than that observed in normal subjects after to measure adrenal cortical function by measuring the continuous corticotropin infusion (46, 49). the urinary cortisol or corticosterone concentration. Table V shows the data on the miscible pool size Furthermore, the urine cortisol and corticosterone and turnover rate of corticosterone in three pa- levels are essentially a reflection of the plasma tients with cirrhosis. level and may not be a good index of the rate of secretion of these steroids. DISCUSSION A urine steroid assay procedure capable of measuring all of the metabolites of cortisol would, of The importance of the liver in the regulation and course, serve as a good quantitative assay of control of the level of circulating steroid becomes adrenal cortical function; however, since the uri- apparent when it is realized that the intensity of nary metabolites of the steroids are so numerous, action of the hormone must be under continuous any measurement of a single metabolite or group dynamic control. This rapid control is possible of metabolites may not represent a reliable index because of mechanisms within the liver that act of adrenal secretion rate. A slight change in the to interrupt, by chemical attack, the biologic action pattern of the steroid metabolites may alter any of the hormone. Such enzymatic processes, that interpretations of secretion rates. The phenyl- serve to terminate or to diminish the action of the hydrazine colorimetric assay procedures after glu- steroid through conversion to inactive products, curonidase hydrolysis measure only 20 to 30 per may be subject to alteration by diseases, especially cent of the total metabolites of cortisol, approxi- those affecting the liver. mately one-half of the steroid metabolites that are Previously, many of the conclusions regarding conjugated with glucuronic acid (15, 55). Such adrenal function in liver disease were drawn from an assay, however, may serve as an index of the studies of the urinary 17-ketosteroids which were relative state of activity of the adrenal cortex. usually found to be low. The patients in the pres- The results obtained in this study and the stud- ent series were also found to have low urinary ies of Brown and associates (32), who used simi- Adrenal function has also been 17-ketosteroids. lar methods for assay of urinary corticosteroids, evaluated on the basis of studies of urinary corticosteroids; however, in many instances rather non- demonstrated that the urinary metabolites of corti- specific assay methods were used. Frequently the sol were below normal. These investigators (32) assay procedures did not utilize a hydrolytic pro- also found that patients with liver disease and cedure. Acidification of the urine to pH 1 (53) normal subjects excreted the same quantity of an does not hydrolyze the metabolites of cortisol (15). infused dose of cortisol as phenylhydrazine-react- Since less than one per cent of the total quantity ing steroid metabolites conjugated with glucuronic of cortisol or corticosterone synthesized per day acid (Table III). If patients with liver disease is excreted in the urine (15, 54), it is not possible and normal subjects excreted a comparable frac- ADRENAL FUNCTION IN LIVER DISEASE 327 tion of endogenously synthesized cortisol as phenyl- fused cortisol from the plasma. This defect in the hydrazine-reacting metabolites, then a low urine capacity of the liver to metabolize cortisol was corticosteroid level in patients with cirrhosis might quite specific since other naturally occurring ster- suggest that adrenal cortisol secretion would be de- oids were metabolized at a rate close to normal. creased. The data on urine corticoid concentration This was true for the biologically active corticoster- (Table I) and rate of synthesis of endogenous cor- one and cortisone, the biologically active synthetic tisol (milligrams per day, Table IV) might appear 9a-fluorocortisol and Al 9a-fluorocortisol, and the to be inconsistent with the data presented in Table biologically inactive steroids 20a- and 20,3-cortisol III (e.g., in Patient T.T., both the urine corticoid (15), dihydrocortisone, dihydrocortisol, tetrahy- level and cortisol turnover were 1.9 mg per day). drocortisone, and tetrahydrocortisol. The rate of The poor specificity of the urine corticoid assay metabolism of aldosterone also did not appear to in this patient probably accounts for the fact that be markedly impaired. these two figures are identical (see also Patients When infused rapidly in quantities in the range D.R. and M.B.). In normal subjects with normal of 100 mg, most steroids disappeared at rates ap- urine corticoid levels the urine corticoids usually proximately proportional to their concentration in account for 30 to 50 per cent of the total daily peripheral blood plasma. If the steroid is dis- corticoid production (49). tributed within the plasma and body tissues and if In all of the subjects with liver disease, the there is a state approximating equilibrium between plasma cortisol level was within the normal (8:00 plasma and the body tissues, the plasma concen- a.m. range). The specificity of this determination tration may be expressed as a simple exponential was evaluated both in patients with hepatitis and function of time. For comparisons of steroids with cirrhosis by the previously described isotope with respect to their rate of metabolism or for dilution method. This evaluation indicated that purposes of comparing the rates of metabolism of 80 to 95 per cent of the material measured with steroids in different diseases, it is convenient to the phenylhydrazine steroid assay method was apply the term "biologic half-life" to this expo- cortisol. Plasma corticosterone levels were slightly nential process; i.e., the time at which half of any lower than normal, or at the lower limit of normal. given quantity remains in the plasma. After the The presence of a normal concentration of cortisol rapid infusion of a trace (microgram) quantity in the plasma indicates that the patients with liver of the steroids cortisol or corticosterone, the bio- disease had no deficiency of circulating cortisol. logic half-life was shorter despite the fact that, with A normal level of cortisol can, however, be main- both the microgram and milligram doses, the de- tained in the plasma by either a decreased rate of cline in plasma concentration represented a first catabolism of cortisol and an associated decreased order process. These differences in rate of disap- rate of synthesis, or by an increased synthesis and pearance of steroid from the plasma may result an increased metabolism of the steroid. Thus, a from the fact that equilibration between plasma normal plasma steroid level establishes that corti- and tissues is not approximated, and the plasma sol is being synthesized, but fails to indicate its rate concentration may be expressed not as a single of synthesis. exponential but rather as a series of exponential Differences in fraction of bound and unbound terms in which the exponents are functions of the cortisol (56-59) may occur in liver disease. Re- rate constants for chemical transformation within cently it has been reported that some patients with the liver, diffusion of the steroid from the plasma liver disease may have an increased fraction of un- into the tissues, and re-entry of the steroid from bound cortisol, yet the total concentration of the tissues into the plasma. A larger fraction of plasma cortisol (bound plus unbound) may be nonprotein-bound steroid available for diffusion normal (60). Sandberg and Slaunwhite (61) into the tissues and re-entry into the plasma fol- however, recently reported that the transcortin lowing the infusion of pharmacologic doses of ster- binding of cortisol was normal in 16 patients with oid might be expected to alter the relative contribu- cirrhosis. tions of these various first order processes. That All of the patients with liver disease showed a this may be so is suggested from the fact that with (lecreased rate of clearance (metabolism) of in- the larger doses of cortisol and corticosterone, a 328 RALPH E. PETERSON three- to fivefold larger volume of distribution was impaired capacity of the liver to conjugate the ster- obtained (15, 16, 46, 50). The more rapid entry oid metabolites. Also, Patient T.T. with cirrhosis, of labeled steroid into the ascitic fluid in the cir- excreted phenylhydrazine-reacting steroid glucu- rhotic demonstrated that the body tissues and the ronide metabolites (Figure 3) at the same rate and extravascular fluid take up a larger fraction of the in the same quantity as the normal subjects. The infused steroid when it is given in therapeutic patient with congenital nonhemolytic jaundice, quantities. The data from the studies of the with the defective hepatic enzyme system for the plasma protein binding of these two steroids also conversion of bilirubin to its glucuronide conju- lend support to this concept. With the trace dose, gate, showed no defect in his capacity to metabolize close to 100 per cent is protein-bound, whereas infused cortisol (52); however, the rate of steroid with larger doses only 70 to 80 per cent is pro- glucuronide formation was reduced after adminis- tein-bound (56-59). tration of a loading dose of one of the tetrahydro The great difference in the rate of metabolism metabolites of cortisol. of cortisol in the normal subject and the cirrhotic, The slower rate of metabolism of cortisol in the as contrasted to the rather unimportant differences subjects with cirrhosis undoubtedly accounted for in the metabolism of other infused steroids, implies a larger quantity of the infused cortisol in the urine that the slower rate of metabolism of cortisol was because the plasma concentration was maintained not the result of impaired circulation through the at elevated levels for a longer time. liver, as has been suggested for the mechanism of The results of the turnover rate studies utilizing the delayed removal of sulfobromophthalein (32). cortisol-4-C14 are consistent with the plasma and Because of the selective nature of the defect in cor- urine steroid assay data and the studies of the rate tisol metabolism it would seem unlikely that a defi- of metabolism of infused steroids. The miscible ciency of a cofactor such as reduced TPN could be pools of cortisol and corticosterone were normal involved, since TPNH is required for the metabo- or only slightly increased. The figures for the lism of most of the steroids. It seems more reason- miscible pool size, however, probably represent able to propose a defect in the enzyme system (44) only the rapidly turned over and rapidly exchange- for transforming cortisol to dihydrocortisol (5a- or able miscible pool. In the patients with ascites, 5,8-dihydrocortisol dehydrogenase). This repre- the infused cortisol-4-C14 did not mix rapidly with sents one of the predominant initial steps in the the cortisol in the ascitic fluid. chemical inactivation of cortisol and would ap- The patients with liver disease synthesized cor- pear to be the enzyme system most likely involved. tisol and corticosterone at a decreased rate, and Dihydrocortisol was metabolized to tetrahydro- excreted a smaller quantity of metabolites of en- cortisol at a normal rate. This presumes that the dogenous cortisol. These patients thus had transformation of cortisol to dihydrocortisol is a adrenal cortical insufficiency in terms of quantity major pathway for the metabolism of cortisol. It of cortisol synthesized; however, they were es- is not known if the enzymatic defect involves the sentially "eucorticoid" in terms of a plasma cortisol soluble 5,f or the particulate 5a dehydrogenase; level maintained in the normal range by virtue of however, there is evidence that both of these en- a defect in the capacity of the liver to metabolize zymes have a high degree of substrate specificity the circulating cortisol at the normal rate. (62). The primary defect in cirrhosis may be a de- M\'ore than one-half of the metabolites of cortisol creased rate of hepatic enzymatic transformation of have the reduced A ring; however, only about the circulating cortisol (cortisol -> dihydrocortisol) one-half of these are present as C-20-keto and ap- with the consequent alterations in plasma cortisol proximately one-half appear in the urine as the resulting from this disturbance. On the basis of C-20 reduced compounds (55). Thus, it is pos- the mechanism of the negative feedback control of sible that there may also be a defect in the C-20 adrenocortical secretion, changes in the plasma cor- reductase enzyme system. tisol produce opposite changes in the rate of secre- The normal rate of disappearance of infused tion of the adrenocorticosteroids (63). Thus, if tetrahydrocortisol from the plasma demonstrated the rate of inactivation of cortisol is diminished, the that in cirrhosis of the liver there is no markedly increment in plasma cortisol will inhibit cortico- ADRENAL FUNCTION IN LIVER DISEASE 329 tropin secretion, and the resultant depression of increase in the urinary aldosterone level. In a re- adrenal cortisol synthesis will serve to maintain a cent report of a case of cirrhosis, the aldosterone normal concentration of plasma cortisol. Urqu- secretion rate was found to be 1,500 to 2,000 ,ug hart, Yates and Herbst (64) have shown a pri- per day and did not appear to be influenced by mary diminution of hepatic capacity for adreno- changes in sodium intake (70). Also, unlike cortical steroid inactivation in the rat after surgical cortisol metabolism, these changes in aldosterone removal of more than one-half of the liver. Fol- metabolism were not associated with a striking lowing partial hepatectomy secondary adrenal change in its rate of metabolism. atrophy occurred. The liver has also been presumed to play an important role in the regulation SUMMARY of adrenal cortical activity in thyroid disease (65). Plasma cortisol and corticosterone levels were If the urinary 17-ketosteroid content is a reas- normal in patients with liver disease; however, onably reliable index of the production of 17- urine corticoids and 17-ketosteroids were usually ketosteroids by the adrenal cortex, then the data below the normal level. Cortisol was metabolized from this and other studies would seem to indicate at a diminished rate in the cirrhotics; however, that adrenal 17-ketosteroid production as well as most other infused steroids were metabolized at a cortisol and corticosterone secretion are depressed. normal rate. This defect in the metabolism of cor- The low urine 17-ketosteroid levels are probably tisol was discussed in terms of a low hepatic ac- not the result of a decreased conversion of corti- tivity of the specific TPNH-dependent enzyme costeroid to 17-ketosteroids since normally the catalyzing the reduction of the 4,5 double bond. 17-ketosteroid metabolites of cortisol make up less No impaired conjugation of the tetrahydrosteroid than 10 per cent of the cortisol metabolites, and metabolites of cortisol was demonstrated. The thus only a small fraction of the total urine 17- rate of synthesis of cortisol and corticosterone was ketosteroids (28, 66). markedly reduced in the patients with cirrhosis. Whereas adrenal glucocorticoid (cortisol and The rate of synthesis of aldosterone was increased corticosterone) and adrenal androgen (17-keto- in patients with cirrhosis maintained on a low so- steroids) secretion are frequently suppressed in dium intake. Patients with cirrhosis have a rela- liver disease, mineralocorticoid (aldosterone) tive adrenal insufficiency in the secretion of corti- production appears not to be decreased. Urine sol and corticosterone although under basal con- aldosterone levels are known to be either normal ditions they maintain a normal plasma cortisol or increased in hepatic cirrhosis (67-69). In pa- level because they metabolize cortisol at a slower tients with ascites, the urine aldosterone levels are rate. These data suggest that there is a homeo- usually markedly elevated; however, in patients static mechanism mediated through the liver- without ascites, normal levels are usually found. pituitary-adrenals which results in a decreased The urine aldosterone concentration, like the urine synthesis of cortisol and corticosterone in patients corticoid level, may not always adequately reflect with liver disease in whom the rate of removal of the secretory activity of the adrenal cortices, since cortisol by the liver is impaired. the usual assay procedure for urinary aldosterone measures less than 5 per cent of the secreted steroid ACKNOWLEDGMENT (41). Turnover rate studies on six patients with We are indebted to Dr. Rudi Schmid for permission to cirrhosis maintained on 200 to 500 mg sodium study Patient J.D., with familial nonhemolytic icterus. diets yielded markedly elevated rates of secretion We are also indebted to the Upjohn, Merck, and Schering pharmaceutical companies for generous supplies of steroids. of aldosterone- 500 to 3,000 jug per day (41) .4 In all subjects urine aldosterone levels were also ele- REFERENCES vated; however, the increase in aldosterone se- 1. Zondek, B. Vber das Schicksal des Follikelhormons cretion was usually proportionally greater than the (Follikulin) im Organismus. Skand. Arch. Phys- iol. 1934, 70, 133. 4 In six normal ambulatory subjects on an ad lib. so- 2. Samuels, L. T., and West, C. D. The intermediary dium intake aldosterone secretion rates ranged from 150 to metabolism of the non-benzenoid steroid hormones. 400 ,ug per day. Vitam. and Horm. 1952, 10, 251. 330 RALPH E. PETERSON

3. Schneider, J. J., and Horstmann, P. M. Effects of 19. Gilder, H., and Hoagland, C. L. Urinary excre- incubating Compound E and related steroids with tion of estrogens and 17-ketosteroids in young, various surviving rat tissues. J. biol. Chem. 1952, adult males with infectious hepatitis. Proc. Soc. 196, 629. exp. Biol. (N. Y.) 1946, 61, 62. 4. Louchart, J., and Jailer, J. W. Inactivation of 11- 20. Eisenberg, H. L., Kirshen, M. M., Atlas, D. H., and dehydro-17-hydroxycorticosterone by tissue slices. Gaberman, P. 17-Ketosteroid excretion in liver Proc. Soc. exp. Biol. (N. Y.) 1952, 79, 393. disease. Gastroenterology 1951, 18, 36. 5. Berliner, D. L., Grosser, B. I., and Dougherty, T. F. 21. Peterson, R. E., Guerra, S., and Sborov, V. M. The metabolism of cortisol in eviscerated rats. Steroid excretion patterns in acute viral hepa- Arch. Biochem. 1958, 77, 81. titis, with and without adrenocorticotrophin in- 6. Robbins, E. D., Burton, S. D., Byers, S. O., Friedman, fusion. J. Lab. clin. Med. 1954, 43, 58. M., St. George, S., and Ishida, T. Hydrocorti- 22 Dohan, F. C., Richardson, E. M., Bluemle, L. W., Jr., sone metabolism in the perfused isolated rat liver. and Gyorgy, P. Hormone excretion in liver dis- J. clin. Endocr. 1957, 17, 111. ease. J. clin. Invest. 1952, 31, 481. 7. Miller, L. L., and Axelrod, L. R. Cortisone metabo- 23. Bongiovanni, A. M., and Eisenmenger, W. J. Adrenal lism in the perfused normal and experimental cir- cortical metabolism in chronic liver disease. J. rhotic rat liver. Metabolism 1954, 3, 438. clin. Endocr. 1951, 11, 152. 8. Caspi, E., and Hechter, 0. Corticosteroid metabo- 24. Ahrens, E. H., Payne, M. A., Kunkel, H. G., Eisen- lism in liver. IV. Isolation of certain cortisol-4-C1' menger, W. J., and Blondheim, S. H. Primary metabolites. Arch. Biochem. 1956, 61, 299. biliary cirrhosis. Medicine (Baltimore) 1950, 29, 9. Hechter, O., Frank, E., Caspi, E., and Frank, H. 299. Corticosteroid metabolism in liver: In vivo me- 25. Sprague, R. G., Mason, H. L., and Power, M. H. tabolism of cortisone and cortisol by dog liver. Physiologic effects of cortisone and ACTH in man. Endocrinology 1957, 60, 705. Recent Progr. Hormone Res. 1951, 6, 315. 10. Bradlow, H. L., Dobriner, K., and Gallagher, T. F. 26. Zarrow, M. X., Munson, P. L., and Salter, W. T. A The fate of cortisone-T in mice. Endocrinology comparison of androgens determined biologically 1954, 54, 343. and 17-ketosteroids determined chemically in urine 11. Ulrich, F., and Long, C. N. H. Effect of stress on (normal and abnormal). J. clin. Endocr. 1950, 10, serum C1' levels in rats following administration of 692. hydrocortisone-4-C1' and corticosterone-4-C'4. En- 27. Glass, S. J., Edmondson, H. A., and Soll, S. N. Sex docrinology 1956, 59, 170. hormone changes associated with liver disease. 12. Tomizawa, H. H., Narahara, H. T., Gibbons, C. A., Endocrinology 1940, 27, 749. and Williams, R. H. A metabolic study of hydro- 28. Sandberg, A. A., Chang, E., and Slaunwhite, W. R., cortisone in rats. Proc. Soc. exp. Biol. (N. Y.) Jr. Conversion of 4-C14-cortisol to C'4-17-keto- 1954, 85, 51. steroids. J. clin. Endocr. 1957, 17, 437. 13. Nelson, D. H., Samuels, L. T., Williardson, D. G., 29. Shadaksharappa, K., Calloway, N. O., Kyle, R. H., and Tyler, F. H. The levels of 17-hydroxycorti- and Keeton, R. W. Excretion of steroidal sub- costeroids in peripheral blood of human subjects. stances by the adrenal cortex in various diseases. J. clin. Endocr. 1951, 11, 1021. J. clin. Endocr. 1951, 11, 1383. 14. Plager, J. E., Samuels, L. T., Ballard, A., Tyler, F. 30. Goldman, R., and Bassett, S. H. Diurnal variation in H., and Hecht, H. H. The rate of metabolism of the urinary excretion of neutral lipid-soluble re- cortisol in a normal human man. J. clin. Endocr. ducing steroids in congestive cardiac failure and 1957, 17, 1. cirrhosis of the liver with ascites. 15. Peterson, R. E., Wyngaarden, J. B., Guerra, S. L., J. clin. Invest. Brodie, B. B., and Bunim, J. J. The physiological 1952, 31, 253. disposition and metabolic fate of hydrocortisone 31. Schedl, H. P., Ditto, K., and Bean, W. B. Cortico- in man. J. clin. Invest. 1955, 34, 1779. steroid excretion in liver disease. J. Lab. clin. Med. 16. Migeon, C. J., Sandberg, A. A., Decker, H. A., Smith, 1953, 42, 116. D. F., Paul, A. C., and Samuels, L. T. Metabo- 32. Brown, H., Willardson, D. G., Samuels, L. T., and lism of 4-C14-cortisol in man: Body distribution Tyler, F. H. 17-Hydroxycorticosteroid metabolism and rates of conjugation. J. clin. Endocr. 1956, 16, in liver disease. J. clin. Invest. 1954, 33, 1524. 1137. 33. Englert, E., Jr., Brown, H., Wallach, S., and Simons, 17. Hellman, L., Bradlow, H. L., Adesman, J., Fuku- E. L. Metabolism of free and conjugated 17-hy- shima, D. K., Kulp, J. L., and Gallagher, T. F. droxycorticosteroids in subjects with liver disease. The fate of hydrocortisone-4-C1' in man. J. clin. J. clin. Endocr. 1957, 17, 1395. Invest. 1954, 33, 1106. 34. Wallace, E. Z., Christy, N. P., and Jailer, J. W. 18. Fraser, R. W., Forbes, A. P., Albright, F., Sulko- Clinical application of the simplified Silber-Porter witch, H., and Reifenstein, E. C., Jr. Colori- method for determining plasma 17-hydroxycorti- metric assay of 17-ketosteroids in urine. J. clin. costeroids. J. clin. Endocr. 1955, 15, 1073. Endocr. 1941, 1, 234. 35. Lloyd, C. W., and Williams, R. H. Endocrine ADRENAL FUNCTION IN LIVER DISEASE 331

changes associated with Laennec's cirrhosis of the sterone, cortisol, and corticosterone in biological ex- liver. Amer. J. Med. 1948, 4, 315. tracts, particularly applied to human urine. Bio- 36. Barr, R. W., and Sommers, S. C. Endocrine abnor- chem. J. 1957, 65, 639. malities accompanying hepatic cirrhosis and hepa- 55. Fukushima, D. K., Leeds, N. S., Bradlow, H. L., toma. J. clin. Endocr. 1957, 17, 1017. Kritchevsky, T. H., Stokem, M. B., and Gallagher, 37. Peterson, R. E., Karrer, A., and Guerra, S. L. Eval- T. F. The characterization of four new metabo- uation of the Silber-Porter procedure for deter- lites of adrenocortical hormones. J. biol. Chem. mination of plasma hydrocortisone. Analyt. Chem. 1955, 212, 449. 1957, 29, 144. 56. Sandberg, A. A., Slaunwhite, W. R., Jr., and An- 38. Silber, R. H., and Porter, C. C. The determination toniades, H. N. The binding of steroids and steroid of 17,21-dihydroxy-20-ketosteroids in urine and conjugates to human plasma proteins. Recent plasma. J. biol. Chem. 1954, 210, 923. Progr. Hormone Res. 1957, 13, 209. 39. Holtorff, A. F., and Koch, F. C. The colorimetric 57. Daughaday, W. H. Binding of corticosteroids by estimation of 17-ketosteroids and their application plasma protein. III. The binding of corticosteroids to urine extracts. J. biol. Chem. 1940, 135, 377. and related hormones by plasma and plasma pro- 40. Peterson, R. E. The identification of corticosterone tein fraction as measured by equilibrium dialysis. in human plasma and its assay by isotope dilution. J. clin. Invest. 1958, 37, 511. J. biol. Chem. 1957, 225, 25. 58. Bush, I. E. The physiochemical state of cortisol in 41. Kliman, B. and Peterson, R. E. Unpublished obser- blood. Ciba Found. Coll. Endocr. 1957, 11, 263. vations. 59. Slaunwhite, W. R., and Sandberg, A. A. Transcor- 42. Peterson, R. E. The application of tritium to the tin: A corticosteroid-binding protein of plasma. assay of steroids in biological extracts in Proc. of J. clin. Invest. 1959, 38, 384. the Symposium on Advances in Tracer Applications 60. Daughaday, W. H. Binding of corticosteroids by of Tritium. Boston, New England Nuclear Corp. plasma proteins. Arch. intern. Med. 1958, 101, 286. 1959, p. 16. 61. Sandberg, A. A., and Slaunwhite, W. R. Trans- 43. Peterson, R. E., Pierce, C. E., Wyngaarden, J. B., cortin: A corticosteroid-binding protein of plasma. Bunim, J. J., and Brodie, B. B. The physiological II. Levels in various conditions and the effects of disposition and metabolic fate of cortisone in man. estrogens. J. clin. Invest. 1959, 38, 1290. J. clin. Invest. 1957, 36, 1301. 62. McGuire, J. S., and Tomkins, G. M. The multiplicity 44. Tomkins, G. M. Enzymatic mechanisms of hormone and specificity of A4-3-ketosteroid hydrogenases metabolism. I. Oxidation-reduction of the steroid (5a). Arch. Biochem. 1959, 82, 476. nucleus. Recent Progr. Hormone Res. 1956, 12, 63. Sayers, G., and Sayers, M. A. The pituitary-adrenal 125. system. Recent Progr. Hormone Res. 1948, 2, 81. 45. Wilzbach, K. E. Tritium-labeling by exposure of 64. Urquhart, J., Yates, F. E., and Herbst, A. L. He- organic compounds to tritium gas. J. Amer. chem. patic regulation of adrenal cortical function. En- Soc. 1957, 79, 1013. docrinology 1959, 64, 816. 46. Peterson, R. E., and Wyngaarden, J. B. The mis- 65. Peterson, R. E. The influence of the thyroid on cible pool and turnover rate of hydrocortisone in adrenal cortical function. J. clin. Invest. 1958, 37, man. J. clin. Invest. 1956, 35, 552. 736. Pearlman, W. H. [16-H] Progesterone metabolism 47. 66. Dorfman, R. I. In vivo metabolism of neutral steroid in advanced pregnancy and oophorectomized-hys- Endocr. 14, 318. terectomized women. Biochem. J. 1957, 67, 1. hormones. J. clin. 1954, 48. Cope, C. L., and Black, E. The production rate of 67. Axelrad, B. J., Cates, J. E., Johnson, B. B., and cortisol in man. Brit. med. J. 1958, 1, 1020. Luetscher, J. A., Jr. Aldosterone in urine of nor- 49. Peterson, R. E. The miscible pool and turnover rate mal men and of patients with oedema. Its in- of adrenocortical steroids in man. Recent Progr. creased recovery after hydrolysis with acid and Hormone Res. 1959, 15, 231. with beta-glucoronidase. Brit. med. J. 1955, 1, 50. Samuels, L. T., Brown, H., Eik-Nes, K., Tyler, F. H., 196. and Dominguez, 0. V. Extra-adrenal factors af- 68. Wolff, H. P., and Koczorek, K. R. Ueber die Aus- fecting the levels of 17-hydroxycorticosteroids in scheidung von Aldosteron im Harn von Patienten plasma. Ciba Found. Coll. Endocr. 1957, 11, 208. mit Hepatitis Epidemica und Lebercirrhose. Klin. 51. Schmid, R. Some aspects of bile pigment metabo- Wschr. 1955, 33, 1104. lism. Clin. Chem. 1957, 3, 394. 69. Dyrenfurth, I., Stacey, C. H., Beck, J. C., and Ven- 52. Peterson, R. E., and Schmid, R. A clinical syn- ning, E. H. Aldosterone excretion in patients drome associated with a defect in steroid glucuro- with cirrhosis of the liver. Metabolism 1957, 6, nide formation. J. clin. Endocr. 1957, 17, 485. 544. 53. Klein, R., Papadatos, C., Fortunato, J., Byers, C., and 70. Ulick, S., Laragh, J. H., and Lieberman, S. The iso- Puntereri, A. Serum corticoids in liver disease. lation of a urinary metabolite of aldosterone and J. clin. Endocr. 1955, 15, 943. its use to measure the rate of secretion of aldo- 54. Ayres, P. J., Garrod, O., Simpson, S. A., and Tait, sterone by the adrenal cortex in man. Trans. Ass. J. F. A method for the determination of aldo- Amer. Phycns 1958, 71, 225.